Gemini 4 Argon ingests 6M-token MRI series and outputs 26 interactive 3D annotations
clmt · x · 2026-10-29
A user ran a full left-shoulder MRI — all 8 sequences, 6M tokens of context — through Gemini 4 Argon, which produced 26 interactive 3D callouts mapping both the tears and every intact structure in the joint.
The demo highlights what ultra-long-context models could do for medical imaging: joint analysis across all sequences rather than reading slices one by one.
More from Models
- Mistral open-weights Le Chonk: 1T-parameter multimodal model leads open benchmarks in coding and defensive security — _AndrewZhao · 2026-10-30
- Google admits AI search monetizes worse than organic, stock drops ~12% on earnings — Ars Technica AI · 2026-10-30
- Grok switches to Arabic mid-analysis, blames an en-AR language glitch — dbasch · 2026-10-30
- Ollama roundup: Mistral Large 4, DeepSeek v4.1-flash and more land in two weeks — ollama · 2026-10-30
- Ex-OpenAI safety VP Miles Brundage: Claude Duet as a separate app is not ideal — Miles_Brundage · 2026-10-30
- Phonic 1 Tops Artificial Analysis Speech-to-Speech Index at 87.1, Edging Grok Voice 3.0 — ArtificialAnlys · 2026-10-30